Intel launched a new range of processors on Tuesday, a product line that the company says will mean faster, smaller, cheaper, and more energy-efficient computers for everyone. Hurray! Now if only I could understand what they’re talking about. See, the company calls its new microprocessors the Core i5 and the Core i7. People who pay close attention to Intel might find this confusing; after all, didn’t the company already unveil the Core i7? Yes, it did—the Core i7 actually came out about a year ago, with Intel billing it as “the fastest processor on the planet.” OK, so the new chip is just a faster, better version of that old one, right? Actually, no. The processors that Intel unveiled this week are slower than the Core i7 it put out last year. That’s because Intel is now filling out its “midrange” line. It now sells two completely different kinds of Core i7 chips, and the old ones are more powerful than the new ones, violating the computer-industry axiom that newer is always better.
Don’t worry if none of that makes sense to you. Names for Intel’s microprocessors—the chip that sits at the heart of your PC or Mac—haven’t made sense for a long time. It’s not just that the company offers a distressingly wide range of processor options but that—unless you’ve been following Intel’s development roadmap—there’s no way to tell which chips are better than others. For starters, if you’re in the market for a new machine, you can’t rely on the clock rate—the gigahertz number that people usually take as a shorthand for speed—because a processor’s design affects its performance as much as the rate of its clock (this is known as the megahertz myth). Instead, you’ve got to look at what kind of chip a computer has: For Intel’s line, a computer with a Core Duo processor will generally be slower than one with a Core 2 Duo, which is in turn slower than a machine with a Core 2 Quad. But a Core i5 is better than a Core 2 Quad, though not as good as the Core i7. Over the next year, Intel plans to release the Core i3 and the Core i9, which will fit somewhere along the spectrum of Core 2 Duos and Quads—though don’t ask me how.
This gets to a more basic problem with the way computers are marketed. Whether you’re shopping online or at a store, a computer is presented to you as a blizzard of numbers—2.80 GHz Core 2 Duo, 3 GB RAM memory, 320 GB hard drive (7200 RPM), 512 MB video card, etc. The issue is more obvious on PCs, but it’s true of Macs as well. Some people understand these numbers and can use them to make sophisticated cost-benefit comparisons across different specifications. Most people can’t—and they shouldn’t have to, either. My dad, who is in other ways an accomplished fellow, can’t for the life of him figure out the difference between RAM and the hard drive—after all, it’s all just memory, right? Why does a computer have 3 GB of space in one place and 320 GB elsewhere? And which should you have more of? And if your computer’s full, can you just add more?
This is an incredibly cumbersome way to sell a consumer product, and it’s a mystery why computer makers persist in headlining these figures. I’ve got a better (and pretty obvious) idea for marketing PCs—tell us what the computer can do. Every machine on sale should display a tag that lists a half-dozen or so basic stats describing its capabilities. Off the top of my head, these would include: 1) how many seconds the computer takes to load up popular programs like Internet Explorer, Microsoft Word, and iTunes; 2) how quickly it can load Yahoo Mail on a 5-megabit broadband line; 3) how long it takes to rip a copy of Michael Jackson’s Thriller CD to MP3s; 4) how many photos, songs, and movies you can store on its hard drive; 5) whether it’s capable of playing full-resolution high-definition movies; 6) how fast it can auto-fix a set of 100 snapshots in a popular photo-editing application; 7) and, if it’s a laptop, how long the battery will last given normal usage (as David Pogue has suggested).
I understand these numbers don’t reflect all potential uses. But they at least give customers a sense of the meaningful differences between two computers. A more powerful processor, for instance, is going to make your computer noticeably faster only during certain processor-intensive tasks—if you’re ripping a lot of music, compressing a lot of files, or retouching a lot of photos. Getting a Core i5 rather than a Core 2 Duo machine isn’t likely to make much difference if you spend most of your time browsing the Web, where speed is usually limited by your network connection, not your processor. If all you’re looking for is a fast computer for the Web, you could compare a $300 desktop with a $700 desktop just on point No. 2—and you’d discover that they’re pretty similar.
Of course, that scenario may suggest why computer makers haven’t adopted this common-sensical model for selling machines—most computers on sale today, even very cheap ones, are so powerful that they’d excel at all of my tests. If computer makers were honest about what a $300 computer could do, wouldn’t they lose a lot of business that now comes to them simply out of confusion?
Perhaps. On the other hand, there’s probably a branding upside to clearing up the mystery surrounding PC stats. A manufacturer like Dell, for instance, which isn’t known for its consumer friendliness, could remake itself by coming out as the first PC maker to offer Real-Life Stats™. At the very least, the company would get a shout-out from me.